Autism spectrum disorder (ASD) is characterized by social and communication deficits as well as restricted, repetitive behaviors (RRBs). RRBs include ?lower order? motor stereotypies such as body rocking and hand waving as well as more elaborate, compulsive behaviors or ?higher order? RRBs such as checking and hoarding and insistence on sameness. Although RRBs, along with social and communicative deficits, constitute the triad of symptoms essential to the diagnosis of autism, RRBs have been surprisingly understudied. The extant question in the field is: what neural circuits mediate RRBs and how might these circuits be altered in ASD and thus targeted for treatment? Insistence on sameness behavior can be measured in tasks of cognitive flexibility using reversal learning paradigms. Individuals with ASD have specific deficits in reversal learning paradigms that use a probabilistic reinforcement schedule, which can be tested in mice using similar cross-species paradigms to probe neural circuitry and treatments. Because prenatal inflammation is a risk factor for autism and other neurodevelopmental disorders, we and others have conducted studies of maternal immune activation (MIA) in model organisms. In addition to social deficits, MIA mice have deficits in probabilistic reversal learning similar to that observed in ASD. Disruptions in frontal cortex function, particularly orbito-frontal cortex (OFC) and striatum lead to deficits in cognitive flexibility. Preliminary data from our lab show that OFC->dorsal medial striatum (DMS) circuit activation disrupts probabilistic reversal learning in a manner consistent with ASD. Intriguingly, structural and functional abnormalities of the OFC are evident in ASD, which may contribute to RRBs. Thus, we hypothesize that probabilistic reversal learning deficits in MIA mice are due to overactivation of OFC-DMS circuit and that inhibiting this circuit will ameliorate the deficits. Aim 1 studies will test whether optogenetic activation of OFC->DMS glutamate projections will exacerbate RRBs in MIA mice and/or replicate RRBs in control mice. Channelrhodopsin (ChR2, a blue light-gated cation channel) will be expressed via the CamKII promoter selectively in glutamate neurons in the OFC. Fiber optic probes will be implanted in the DMS to stimulate terminals from the OFC during the reversal phase of the task. Aim 2 studies test whether OFC-DMS circuit inhibition ameliorates cognitive flexibility deficits in MIA mice. Specifically, a CamKII promoter-driven halorhodopsin (eNpHR3.0, a green light-gated chloride ion pump) will be infused into the OFC, and stimulation of fiber optic probes in the DMS will test whether optogenetic inhibition of OFC->DMS neurons attenuates cognitive flexibility deficits in MIA mice. Because both ASD and MIA are associated with decreased GABA function in frontal cortex, Aim 3 studies test whether enhancing OFC GABA transmission ameliorates cognitive flexibility deficits in MIA mice. These experiments will provide evidence for a specific circuit involved in RRBs in MIA and a possible remediation through inhibition of the circuit and potential treatment approaches for RRBs in ASD.